This project will investigate and develop effective information technologies for comparative analysis and visualization of complex data generated by comprehensive two-dimensional gas chromatography (GCxGC). GCxGC is an emerging technology that provides an order-of-magnitude greater separation capacity, significantly better signal-to-noise ratio, and higher dimensional retention-structure relations than traditional GC. The principal challenge for utilization of GCxGC, in a wide range of public-health and other applications, is the difficulty of analyzing and interpreting the large, complex data it generates. The quantity and complexity of GCxGC data necessitates the investigation and development of new information technologies. This project will develop and demonstrate innovative methods and tools for comparative analysis of GCxGC datasets. The expected results of this research and development include a PCA-based method for chemical fingerprinting, decision trees with chemical constraints for sample classification, genetic programming for template and constraint-based matching and classification, and visualization methods for comparative GCxGC analyses. These methods will be implemented in commercial software that will support researchers and laboratory analysts in a wide range of commercial applications, including health care, environmental monitoring, and chemical processing. The power of GCxGC, supported by effective information technologies, will enable better understanding of chemical compositions and processes, a foundation for future scientific advances and discoveries. Relevance to Public Health. Today, a few advanced laboratories are pioneering GCxGC for a variety of applications such as environmental monitoring of exposure profiles in air, soil, food, and water; identification and quantification of toxic products in blood, urine, milk, and breath samples; and qualitative and quantitative metabolomics to provide a holistic view of the biochemical status or biochemical phenotype of an organism. Many analyses in these applications require detailed chemical comparisons of samples, e.g..monitoring changes, comparison to reference standards, chemical matching or "fingerprinting", and classification. GCxGC is a powerful new technology for such comparative analyses. This proposal will provide innovative information technologies to support users in these applications. [unreadable] [unreadable] [unreadable]